Improvement from January 2016 Front Office to Front Line …€¦ · spreading and sustaining improved compliance have proved dif-ficult.11,12 As part of its efforts to reduce HAIs,

www.jcrinc.com

Improvement fromFront Office to Front Line

January 2016Volume 42 Number 1

Sustaining Improvement in Hand Hygiene and Health Care–Associated InfectionsFeaturesInfection Prevention and Control

■■ Using the Targeted Solutions Tool® to Improve Hand Hygiene Compliance Is Associated with Decreased Health Care–Associated Infections

Performance Improvement■■ The Costs of Participating in a Diabetes Quality Improvement Collaborative: Variation Among Five Clinics

Risk Assessment and Event Analysis■■ A Tool for the Concise Analysis of Patient Safety Incidents

Care Processes■■ Factors Associated with Inpatient Thoracentesis Procedure Quality at University Hospitals

Departments■■ To Our Readers

Tool Tutorial■■ Post Event Debriefs: A Commitment to Learning How to Better Care for Patients and Staff

“The data from this study suggest that procedure-specific processes to reduce infections

related to devices such as central lines or ventilators or others,

such as urinary catheters, may not achieve extremely low rates

of infection unless they are accompanied by high rates of hand hygiene compliance.”

—Using the Targeted Solutions Tool® to Improve Hand Hygiene Compliance

Is Associated with Decreased Health Care–Associated Infections

(p. 15)

The Joint Commission Journal on Quality and Patient Safety

Volume 42 Number 1January 20166

In 2006 Memorial Hermann Health System (MHHS) began a journey to become a high reliability organization (HRO).1

HROs maintain high levels of quality and safety over long peri-ods with very few adverse events, despite the potential for large-scale harm.2,3 The ultimate goal for hospitals and health care systems striving for high reliability is zero harm for patients and health care workers.4

Consisting of 12 hospitals and more than 200 ambulatory care organizations, specialty programs, and services in the great-er Houston area, MHHS employs more than 24,000 staff and has more than 5,000 physicians. As part of its journey to high reliability, MHHS aimed to reduce its rates of health care–asso ciated infections (HAIs) and, in 2007, began conducting Robust Process Improvement® (RPI®) projects5 to address HAIs. Although these efforts were effective,1 MHHS believed that lower rates of HAIs were possible by directly addressing hand hygiene compliance,6–8 which had not been done in its previ-ous RPI projects. Although studies have demonstrated that im-proving hand hygiene in hospitals reduces rates of infection,9,10 spreading and sustaining improved compliance have proved dif-ficult.11,12 As part of its efforts to reduce HAIs, MHHS chose to voluntarily participate, as did seven other organizations, in the Joint Commission Center for Transforming Healthcare’s (the Center’s) 2009 inaugural project, which was designed to improve hand hygiene compliance.13 As described by Chassin et al., the project participants systematically identified effective and reliable strategies for collecting hand hygiene compliance data, determined root causes and contributing factors for non-compliance, and implemented specific interventions to address each factor.13 In collaboration with the eight organizations, the Center then leveraged this information to create a step-by-step application for hand hygiene (called the Targeted Solutions Tool® [TST®]), which guides Joint Commission–accredited or-ganizations through an improvement project to increase hand hygiene compliance.14

MHHS also helped test the TST methodology and the even-

Infection Prevention and Control

Using the Targeted Solutions Tool® to Improve Hand Hygiene Compliance Is Associated with Decreased Health Care–Associated InfectionsM. Michael Shabot, MD, FACS, FCCM, FACMI; Mark R. Chassin, MD, FACP, MPP, MPH; Anne-Claire France, PhD,CPHQ, MBB, FACHE; Juan Inurria, FACHE, FABC, CPHQ; Jan Kendrick, MA; Stephen P. Schmaltz, PhD

Article-at-a-Glance Background: In 2010 Memorial Hermann Health System (MHHS) implemented the Joint Commission Center for Transforming Healthcare’s (the Center’s) Web-based Targeted Solutions Tool® (TST®) for improving hand hygiene through-out its 12 hospitals after participating in the Center’s first project on hand hygiene, pilot testing the TST, and achieving significant improvement for each pilot unit. Because hand hygiene is a key contributing factor in health care–associated infections (HAIs), this project was an important part of MHHS’s strategy to eliminate HAIs. Methods: MHHS implemented the TST for hand hygiene in 150 inpatient units in 12 hospitals and conducted a sys-temwide process improvement project from October 2010 through December 2014. The TST enabled MHHS to mea-sure compliance rates, identify reasons for noncompliance, implement tested interventions provided by the TST, and sus-tain the improvements. Data on rates of ICU central line– associated bloodstream infections (CLABSIs) and ventilator- associated pneumonia (VAP) were also collected and analyzed.Results: Based on 31,600 observations (October 2010–May 2011), MHHS’s systemwide hand hygiene compli-ance baseline rate averaged 58.1%. Compliance averaged 84.4% during the “improve” phase (June 2011–November 2012), 94.7% in the first 13 months of the “control phase” (Decem ber 2012–December 2014) and 95.6% in the final 12 months (p < 0.0001 for all comparisons to baseline). Con comitantly, adult ICU CLABSI and VAP rates decreased by 49% (p = 0.024) and 45% (p = 0.045), respectively.Conclusion: MHHS substantially improved hand hygiene compliance in its hospitals and sustained high levels of com-pliance for 25 months following implementation. Adult ICU CLABSI and VAP rates decreased in association with the hand hygiene compliance improvements.

Copyright 2016 The Joint Commission


Volume 42 Number 1January 2016 7

tual Web-based application.14 Four MHHS hospitals had units participate in different stages of TST development: Memorial Hermann The Woodlands Hospital, which participated in the initial hand hygiene project; Memorial Hermann Northeast Hospital and Memorial Hermann Heart & Vascular Institute- Texas Medical Center, both of which piloted the implemen-tation of solutions from the original project as the TST was being developed; and Memorial Hermann Northwest Hospital, which piloted the Web-based TST application. All pilot units in the four hospitals achieved significant improvements in hand hygiene compliance while participating in the project. For ex-ample, The Woodlands’ piloting units increased hand hygiene compliance from 27% to 80% during the initial test phase of the Center’s inaugural hand hygiene project; the other three participating units achieved similar results. Because of improve-ments achieved by the four piloting hospitals’ units, MHHS chose to implement the TST for hand hygiene in all 12 of its hospitals when it became available in September 2010.

MHHS hospitals comprise approximately 3,225 total beds, of which 659 (20%) are intensive care beds. The 12 hospitals span a wide range of services from a small (81-bed) commu-nity hospital to medium (250–290 beds) and large (420–560 beds) community hospitals. MHHS’s large tertiary care hospi-tal houses a Level 1 trauma center, a burn center, as well as car-diac, neurosurgical, and transplant (heart, liver, pancreas, and kidney) services. MHHS also includes a children’s hospital and two rehabilitation hospitals. During the study period, the vast majority of MHHS’s inpatient units were comprised of sin-gle-bedded rooms. A small number of two-bed rooms existed in 1 of the 12 hospitals, and these were converted to single-bedded rooms during the study period.

In this article, we describe how MHHS spread the imple-mentation of the hand hygiene TST to all its hospitals. We hy-pothesized that the TST would enable substantial increases in hand hygiene compliance across the system and facilitate the maintenance of high levels of performance following the imple-mentation of improvement interventions. In addition, MHHS measured rates of ICU central line–associated bloodstream infections (CLABSIs) and ventilator-associated pneumonia (VAP) in all its hospitals to assess the relationship between im-proving hand hygiene compliance and these HAIs. Previous evaluations of the TST’s impact on hand hygiene compliance have not assessed any association with rates of HAIs.14 We hy-pothesized that improvements in hand hygiene compliance, if substantial enough, would be associated with declines in the rates of these HAIs.

MethodsThe TST for hand hygiene

The TST for hand hygiene is a systematic, Web-based applica-tion founded on RPI methodologies, which include Lean, Six Sigma, and change management, that guides health care organi-zations through a series of steps to increase hand hygiene com-pliance; it has been described in detail elsewhere.14,15 The TST mirrors the five stages of a Six Sigma project (Define, Measure, Analyze, Improve, and Control [DMAIC]). In the next sec-tion, we describe the specific actions that MHHS took in im-plementing the TST. For purposes of reporting results, we have divided the study’s time line into three main periods, as follows:

1. The “baseline” period included the collection of data on hand hygiene compliance before the implementation of im-provement interventions.

2. The “improve” phase included the time during which in-terventions were implemented.

3. The “control” phase, which followed the deployment of specific improvement actions, was intended to assess whether MHHS was able to sustain any improvements realized during the improve phase.

Although we describe the primary activities that took place during each of these periods, as a practical matter, we know that some improvements were initiated toward the end of the base-line period and that some improvements continued into the control phase. In the first three years that the hand hygiene TST was available, organizations that used it improved compliance from a baseline average of 57.9% to 83.5% (p < 0.0001).14,16

Leading and Managing The TST ProjecTS aT MeMoriaL herMann heaLTh SySTeM

MHHS’s hand hygiene initiative had a project lead [A-C.F.] who was the master black belt (MBB) for quality, patient safety, and infection control, and who also led the initial hand hygiene project and subsequent pilots through the Center. At each hospi-tal, the chief nursing officer (CNO) served as executive sponsor and was responsible for identifying process owners at the facility and unit levels. Most of the process owners primarily held nurs-ing leadership positions; some of the members held respiratory therapy and infection prevention positions. Several of these members had also participated in the Center’s inaugural hand hygiene project or subsequent pilots and then continued on during the larger initiative. The MBB provided on-site support as needed and virtual support via biweekly conference calls for the process owners and other team members who wished to par-ticipate. During the first year of baseline data and contributing factors collection, the project required 50% of the MBB’s time.




Each hospital has a hand hygiene team composed of the fa-cility process owner and a subset of unit process owners. The CNO at each hospital also served as executive sponsor driving the local initiative with the MBB’s help. MHHS began TST projects in three waves or rounds because of the number of hos-pitals and units implementing the TST and the coordination involved for its system (Appendix 1, available in online article). Because local leaderships were more familiar with their hospitals than was the MBB, leadership at each site determined which units participated in each wave, and all 150 units were eligi-ble to use all three rounds, as necessary, to complete training and implementation. Included in these waves also were the two units from The Woodlands and one unit from each of the three hospitals that participated in the Center’s initial hand hygiene project and subsequent pilots.

Leveraging The TST To conducT a SySTeMwide hand hygiene iMProveMenT ProjecT

MHHS progressed through its improvement project between October 2010 and December 2014 to increase hand hygiene compliance by collecting reliable data, identifying compliance barriers, implementing proven intervention solutions to address barriers, and sustaining these interventions, as represented in the six steps, which we now describe.

Step 1. Getting Started. During this step, the executive spon-sors for each hospital and participating unit identified the lead process owners. To ensure accountability at the operational unit level, all process owners were operational hospital managers, directors, or executives rather than members of the quality or infection control departments. With the exception of North-east Hospital, which included an infection preventionist as a process owner because of her previous participation in the hos-pital’s initial pilot project, the only criterion set for determining the lead process owners was that they were not to be infection preventionists.

With the support of senior leadership, MHHS’s quality, pa-tient safety, and infection control leadership and the project lead determined the project’s objectives and goals and then es-tablished the following time line and goals:

■■ Collect reliable data and identify contributing factors (Octo ber 2010 through May 2011).

■■ Determine solutions to address contributing factors and implement solutions (June 2011 through November 2012).

■■ Increase hand hygiene compliance in each unit by 30 per-centage points above each unit’s baseline or to 90% by one year following the implementation of solutions.

On the basis of results achieved by the piloting units, MHHS

selected these goals (1) with the knowledge that an increased compliance rate of 80% to 90% was possible for most of its units and (2) with the expectation that the units with lower baselines could be motivated to achieving much higher rates of compliance. To organize and manage the rollout, a key spread-sheet of monthly performance goals was developed and provid-ed to all lead process owners to facilitate efforts in their units (Appendix 1).

Step 2. Training Secret Observers and Just-in-Time (JIT) Coaches. To collect reliable data, the TST methodology engag-es anonymous secret observers who hold unbiased positions and are able to observe hand hygiene compliance.17–19 These ob-servers—whose identities are completely unknown to the indi-viduals being observed—do not interfere with the caregivers’ work. Their role is to observe and collect data. At MHHS, se-lected staff members were trained on accurate data collection by accessing modules in the TST that required watching train-ing videos and passing a certification examination (score of at least 90%). We did not calculate formal rates of interrater re-liability among secret observers. As it became necessary to de-ploy additional secret observers during the course of the project (for example, if their roles became known to unit staff), the same training program and certification examination were used for the new observers. The secret observers were also trained to identify and record some contributing causes of noncompliance (for example, caregivers’ hands were full of supplies, no alco-hol hand rub dispenser near door to patient room, or caregiver needed several trips in and out of a room to complete a task [fre-quent entry or exit]). However, other contributing causes could not be observed (for example, perception that hand hygiene was not required or perception of skin irritation). Therefore, in ad-dition to the secret observers, the TST also provides training for JIT coaches, who observed noncompliance, obtained infor-mation regarding contributing factors for noncompliance, and provided feedback to noncompliant individuals.7 Secret observ-ers and JIT coaches were trained during training fairs held for participating units to maintain anonymity.

Clinical staff and nonclinical staff, including environmental services, dietary, plant operations, and chaplains, were eligible to participate as secret observers and JIT coaches. Only secret observers’ data were used to calculate rates of noncompliance. When a JIT coach observed a noncompliant individual, he or she intervened to identify contributing factors and root causes for noncompliance and briefly educated the individual about hand hygiene protocol. All JIT coaches were opinion leaders staffed in their respective units and did not serve as secret ob-servers. We recognized that the activities of the JIT coaches




could influence hand hygiene be-havior by themselves. Their work was concentrated in the last three months of the eight-month baseline peri-od (March through May 2011; see Appendix 1.) Because they collect-ed a limited number of observations during a relatively brief time period, we believed that any improvement effect due to those activities would be limited. Therefore, we defined the baseline period as continuing until specific inter ventions began to be de-ployed in June 2011.

Step 3. Measuring Baseline Com- pliance. Between October 2010 and May 2011, all participating units were asked to obtain a collective total of 20 observations per day from se-cret observers for 14 nonconsecutive 24-hour days, including day, night, and weekend shifts, with data record-ed on the Hand Hygiene Observa-tion and Contributing Factor Form (Appendix 2, available in online arti-cle), which was developed entirely by the Center’s collaborative hand hy-giene project and based on measures and identified contributing factors. The observers recorded compliance data on a paper version of the form, printed from the TST. The data were transcribed into the TST database by designated, separately trained individuals. We did not attempt to construct a truly random sample of ob-servations of hand hygiene compliance; nor did we attempt to calculate the percentage of all opportunities to clean hands that was represented in the secret observers’ data.

All caregivers were included in the secret observers’ assess-ments of hand hygiene compliance, including physicians, nurs-es, dietitians, respiratory therapists, phlebotomists, food services workers, and environmental services staff. Only visitors were ex-cluded, as were the rare emergency situations requiring resusci-tation or other emergent interventions. Data were collected on the basis of observations made on entry and exit of a patient’s room and then entered into the TST’s database. For example, an attending physician’s hand hygiene compliance or noncom-pliance on leaving a patient’s room would constitute one obser-vation. By the end of May 2011, MHHS had collected 31,600

observations assessing baseline compliance. All data were en-tered into the TST, which provided real-time data analysis.

Step 4. Determining Factors. During the February through June 2011 time period, each unit collected a minimum of 30 observations (50 observations when possible) to identify con-tributing factors and root causes for noncompliance. We as-sumed that a collection of 30 to 50 observations would be sufficient to reveal which factors most commonly contributed to a unit’s noncompliance. Importantly, the most prominent contributing causes of hand hygiene noncompliance varied from unit to unit and hospital to hospital. Figure 1 (above) displays a strip plot that shows each contributing cause and a distribution of its ranking from 1st to 15th among all causes across the 11 MHHS hospitals included in this analysis. Thus, the most frequent contributing cause (improper use of gloves) ranked first in six hospitals, second in three, and third in two. The next most common (frequent entry or exit) ranked first in two hospitals, second in four, third in two, fourth in one, tied

Variability in Ranking of Contributing Causes of Hand Hygiene Noncompliance Among 11 Memorial Hermann Health System

(MHHS) Hospitals, October 2010–June 2011

Figure 1. The figure displays a strip plot that shows each contributing cause ( ) and a distribution of its ranking from 1st to 15th among all causes across the 11 MHHS hospitals included in this analysis. HH, hand hygiene; meds, medications.

Dispenser broken

Perception of skin irritation

Dispenser empty

Admissions or discharges process

Dispenser location

Isolation area (gown + gloves)

Lack of immediate feedback

Hands full meds

Equipment shared

Perception HH not required

Follow person entry or exit

Distracted or forgot

Hands full supplies

Frequent entry or exit

Improper use of gloves

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15Hospital Rank of Contributing Cause

Con

tribu

ting

Cau

se




for fourth in one, and fifth in one. The substantial variability in the most important contributing causes by hospital was expect-ed and entirely consistent with the experience in the Center’s hand hygiene project and of other hospitals using the TST.13,14 A Pareto chart, which was produced by the TST, was used to graph the frequency of each factor across 11 MHHS hospitals (Figure 2, above). A total of 9,944 observations were represent-ed in the contributing causes of noncompliance contained in the strip plot and Pareto chart. The top six contributing fac-tors identified throughout the system—which accounted for 81% of noncompliance observations—were, in order of average ranking, as follows:

1. Improper use of gloves2. Frequent entry or exit3. Distracted or forgot4. Following a person during entry or exit5. Hands full of supplies6. Ineffective or inconvenient placement of hand gel dis pen-

ser or sink Step 5. Implementing Solutions. Some units began deploy-

ing interventions as early as June 2011. By August 2011, how-

ever, every unit was asked to select targeted solutions that had been test-ed and recommended by the TST to address its spe cific contributing fac-tors.13,14,16 Participating units imple-mented a variety of interventions to address their contributing factors, in-cluding the following:

■ Using a code word to remind a distracted individual to perform hand hygiene (to address “Health care worker was distracted”)

■ Relocating or adding hand gel dispensers to more convenient loca-tions (“Ineffective or inconvenient placement of hand gel dispenser or sink”)

■ Using visual cues to prompt hand hygiene (“Health care worker forgot”)

■ Providing a surface for indi-vid uals carrying meals, supplies, or med i cations to place items and al-low hand hygiene to be performed (“Caregiver’s hands were full . . . no convenient place to put supplies

to facilitate hand hygiene”) Units with similar contributing factors were encour aged to

collaborate. By the end of September 2011, each unit was re-quired to establish a plan for implementing its solutions. By the end of December 2011, each unit was required to implement its solutions.

Each unit had a process owner who spearheaded baseline data collection, identification of the contributing factors to noncompliance, and implementation of targeted solutions us-ing the implementation guides readily available via the TST. Implementation of interventions continued on a rolling basis through November 2012. Thus, we refer to the period from June 2011 through November 2012 as the improve phase of the study, when different interventions continued to be deployed.

Step 6. Sustaining the Gains. In December 2012 MHHS entered the final step of the TST, “Sustaining the Gains,” which is equivalent to the control phase of a Six Sigma project. To sustain the interventions and related improvements, MHHS conducts executive-level monthly operating reviews with each hospital entity to track 60+ quality measures, 1 of which ad-dresses hand hygiene. Each participating unit continues to

Contributing Factors to Hand Hygiene Noncompliance for 11 Memorial Hermann Health System (MHHS) Hospitals,

October 2010–June 2011

Figure 2. A Pareto chart, which was produced by the Targeted Solutions Tool® (TST®), was used to graph the frequency of each factor across 11 MHHS hospitals, with 9,944 observations represented in the contributing causes of noncompliance contained in the strip plot and Pareto chart. HH, hand hygiene.

281 270 251 177 68 15 82,460 1,274 1,227 1,202 1,090 843 496 2823 3 3 2 1 0 025 13 12 12 11 8 5 3

92 95 97 99 100 100 10025 38 50 62 73 81 86 89

Reason

10,000

8,000

6,000

4,000

2,000

0

100

80

60

40

20

Tota

l Cou

nt

Per

cent

Total CountPercent

Cumulative %

Impro

per u

se of

glov

es

Freque

nt en

try or

exit

Distrac

ted or

forgo

t

Follow

perso

n entr

y or e

xit

Hands

full s

uppli

es

Dispen

ser lo

catio

n

Percep

tion H

H not re

quire

d

Lack

of im

mediat

e fee

dbac

k

Isolat

ion ar

ea (g

own +

glov

es)

Equipm

ent s

hared

Hands

full m

eds

Admiss

ions o

r disc

harge

s proc

ess

Dispen

ser e

mpty

Percep

tion o

f skin

irrita

tion

Dispen

ser b

roken




collect and input data accordingly (20 observations of hand hy-giene compliance per month). For all quality measures, if there is a decline in performance, the hospital senior leadership team is expected to address the issues using RPI. In addition, each hospital’s hand hygiene team is responsible for sustaining per-formance. Since December 2012 MHHS has continued to re-cord approximately 10,000 observations per month into the TST. Compliance rates continue to be monitored. If a decrease in compliance occurs, the TST provides for reinstituting the collection of data on contributing causes by secret observers and the JIT coaches to find the reasons for the decrease. As with the initial deployment, the data on contributing causes lead to the selection and implementation of the appropriate targeted solutions. There is no set protocol that defines what level of decreased compliance should trigger these corrective actions. Those decisions are left to each hospital’s hand hygiene team and monitored by leadership.

To acknowledge the accomplishments of individual units and hospitals for reducing rates of specific HAIs to zero, MHHS created the High Reliability Certified Zero Award.1 Between April 2011 and October 2013, 105 Certified Zero Awards were presented to units that did not experience an HAI, patient safe-ty incident, or hospital-acquired condition event for a contin-uous one-year period; 35 of the Zero Awards concerned HAIs.

MeaSuring heaLTh care–aSSociaTed infecTionS

Rates of ICU CLABSI and VAP were measured throughout the study period and for several years before it in a standard-ized manner by certified infection control practitioners using definitions established by the Centers for Disease Control and Prevention (CDC) through its National Healthcare Safety Net-work (NHSN).20 NHSN definitions and measurement pro-cedures were taught and monitored by the MHHS physician epidemiologist throughout the study period. CLABSI rates were measured separately for neonatal ICUs (NICUs) and for adult ICUs. The NHSN definition for VAP was revised begin-ning in 2013, so we were able to include VAP data for analy-sis only through December 2012. We were not able to include other HAIs (for example, catheter-associated urinary tract in-fections and methicillin-resistant Staphylococcus aureus) in this analysis, because data for those HAIs were not collected in the same standardized way during the study period.

STaTiSTicaL MeThodS

To evaluate the impact of the interventions on hand hygiene compliance, we used an interrupted time series analysis with a segmented generalized linear mixed model.21 For purposes of as-

sessing rates of improvement in hand hygiene compliance and HAIs, data from the MHHS hospital that participated in the Center’s first project were excluded because that hospital began its improvements considerably before the baseline data collec-tion period defined in this study for the other 11 MHHS hos-pitals. The model fit separate slopes and intercepts in each of the baseline, improve, and control study phases, allowing the slopes and intercepts within each study period to vary across hospitals. Because we had control phase data for 25 months, we examined the first 13 months (control 1) and the last 12 months (control 2) separately. Least squares means of the fixed effects from this model were then used to calculate the marginal compliance rate within each study phase. We used an F-test to evaluate the ratio of the baseline to the control variance.

We evaluated the impact of the intervention on adult ICU CLABSI, NICU CLABSI, and VAP in several different ways. Because all of these outcomes are rare events, we used Poisson regression techniques.22,23 When the number of these specific infections within hospitals was very small, we aggregated the data over hospitals by month and then analyzed them with a Poisson regression. This method was used to analyze NICU CLABSI (for which only one of six hospitals included in the analysis had more than five of these outcomes over the study period) and VAP (for which only one out of nine hospitals in-cluded in the analysis had more than five outcomes over the study period). For the analysis of ICU CLABSI (for which six of eight hospitals had more than five outcomes), we used a Pois-son hierarchical mixed model to evaluate the covariate trends in outcome rates, using hospital as the random effect. The Poisson regression and Poisson hierarchical mixed model analyses both adjust for the number of patient-days at risk (that is, central line–days for CLABSI and ventilator-days for VAP). The Pois-son hier archical mixed model additionally adjusts for the cor-relation of longitudinal outcome rates within hospital. Because the covariates hand hygiene compliance, time, and study phase are all highly confounded with each other, preventing the eval-uation of all of these variables in a statistical model at the same time, a series of analyses was undertaken to evaluate the impact of each covariate separately. Least squares means were calculated in the analysis using study phase as the covariate to estimate the marginal outcome rates for each study period.

Adult ICU and NICU CLABSI data were available for the entire 51-month study period (October 2010–December 2014), while VAP data were available for only 27 months (October 2010–December 2012). Thus, we were able to evaluate both CLABSI outcomes in all three study phases but VAP data only for the baseline and improve phases. Hospitals were included in




the outcomes analyses if they provided the relevant service. Thus, 8 the 11 hospitals had adult ICUs and were included in the adult ICU CLABSI analysis. The 6 hospitals with NICUs and any central line–days were included in the NICU CLABSI analyses, and the 9 hospitals that treated patients on ventilators were in-cluded in the VAP analyses. All statistical tests were conducted at the 0.05 level of significance.

ResultsiMProveMenT in hand hygiene coMPLiance

A box plot24 displaying monthly hand hygiene compliance rates for all 11 hospitals for all three time periods of the study is shown in Figure 3 (above). Overall, hand hygiene compliance improved by 37.5 percentage points between baseline and the end of the control phase, a relative increase of 65%. Average compliance during the baseline period (October 2010–May 2011) was 58.1%, with a 95% confidence interval (CI) of 42.7% to 72.1%. During the improve phase (June 2011–November 2012), average compliance increased to 84.4%. In the control

phase, compliance averaged 94.7% in the first 13 months and 95.6% in the final 12 months. The generalized linear mixed model demonstrated that the compliance rates in the improve phase and in both parts of the control phase were statistically significant different (p < 0.0001) from the baseline period. The interrupted time series analysis showed that compliance was increasing during the baseline (slope = 0.101, p = 0.026) and improve (slope = 0.078, p = 0.001) phases and stabilized in the control phase (slope = 0.033, p = 0.095). The variability of the compliance rates among hospitals decreased significantly in the control phase (standard error of the mean = 1.6%) compared to baseline (standard error = 7.3%) (p < 0.001).

aSSociaTion of iMProved hand hygiene coMPLiance wiTh decLineS in SeLecTed haiS

Beginning in May 2007—before the current effort to im-prove hand hygiene compliance—MHHS had implemented other interventions to reduce ICU CLABSI and VAP. Those in-terventions focused on care processes specific to each of these

Box Plot of Monthly Hand Hygiene Compliance Rates for 11 Memorial Hermann Health System (MHHS) Hospitals, October 2010–December 2014

Figure 3. The box plot displays monthly hand hygiene compliance rates for 11 MHHS hospitals for all three time periods of the study—baseline, improve, and control.




HAIs, commonly referred to as the central line and VAP “bun-dles”—procedure-specific care processes aimed at reducing the frequency of infection in these clinical situations.25,26 The tra-jectory of the aggregate rates of these HAIs across all MHHS hospitals is shown in Figure 4 (above) for NICU CLABSI, Fig-ure 5 (above, right) for adult ICU CLABSI, and Figure 6 (right) for VAP. We do not attempt here to demonstrate an association between those interventions and the temporal declines in these HAI rates before the current study period. Rather, we present this information to establish the context for the hand hygiene improvement effort. Thus, in these MHHS hospitals a substan-tial effort had been undertaken to implement the CLABSI and VAP bundles before the hand hygiene improvement initiative described here.

We also examined aggregate monthly MHHS data on adult ICU CLABSI and VAP to determine whether the rates of those HAIs were stable in the 12 months—October 2009–September 2010—immediately preceding the study period. Using Poisson regression analysis with time as the covariate, we found no time trend for either adult ICU CLABSI (p = 0.822) or VAP (p = 0.640). Plotting the same data using a control chart with vari-able denominator sizes showed that both rates were well in con-trol and stable with no out-of-control patterns.

Adult ICU CLABSI rates decreased throughout the study period—from 0.83 per 1,000 central line–days at baseline to 0.42 in the last part of control (p = 0.024). VAP rates decreased as well—from 1.04 per 1,000 ventilator–days at baseline to

0.57 during improve (p = 0.045). NICU CLABSI rates did not change significantly. The rate in the baseline period was 1.97 per 1,000 NICU central line–days and 1.84 during con-trol (p = 0.503). These latter events, however, were very un-common, and the present study was not powered sufficiently to detect the magnitude of change observed for the other two outcomes. Power calculations showed that to detect a 50%

Monthly Central Line–Associated Bloodstream Infection (CLABSI) Rates:

All Neonatal ICUs at 7 Memorial Hermann Health System (MHHS) Hospitals,

April 2007–December 2014

0.00

3.00

6.00

9.00

12.00

15.00

CLA

BSI

Rat

e pe

r 1,0

00 L

ine-

Day

s

Reporting Months

Memorial Hermann Healthcare System NICU CLABSIDo No Harm

Central Line Associated Blood Stream Infections

0.00

2.00

4.00

6.00

8.00

10.00

CLA

BSI

Rat

e pe

r 1,0

00 L

ine-

Day

s

Reporting Months

Memorial Hermann Healthcare System Adult ICU CLABSIDo No Harm


8.00

sya

Memorial Hermann Healthcare System Adult VAPDo No Harm

Ventilator Associated Pneumonia

0.00

3.00

6.00

9.00

12.00

15.00

CLA

BSI

Rat

e pe

r 1,0

00 L

ine-

Day

s

Reporting Months

Memorial Hermann Healthcare System NICU CLABSIDo No Harm


0.00

2.00

4.00

6.00

8.00

10.00

CLA

BSI

Rat

e pe

r 1,0

00 L

ine-

Day

s

Reporting Months

Memorial Hermann Healthcare System Adult ICU CLABSIDo No Harm


8.00sya


Ventilator Associated PneumoniaFigure 4. The trajectory of the aggregate rates of the CLABSI rates for the seven MHHS hospitals with neonatal ICUs is shown. Q, quarter.

Monthly Ventilator-Associated Pneumonia (VAP) Rates: All Adult ICUs at

10 Memorial Hermann Health System (MHHS) Hospitals,

January 2006–December 2012

0.00

2.00

4.00

6.00

8.00

CLA

BSI

Rat

e pe

r 1,0

00 V

entil

ator

–Day

s

Reporting Months


Ventilator Associated Pneumonia

Figure 6. The trajectory of the aggregate rates of the VAP rates across the 10 MHHS hospitals with patients on ventilators is shown. Q, quarter.

Monthly Central Line–Associated Bloodstream Infection (CLABSI) Rates: All Adult ICUs at 9 Memorial Hermann

Health System (MHHS) Hospitals, January 2006–March 2015

Figure 5. The trajectory of the aggregate rates of the CLABSI rates for the nine MHHS hospitals with adult ICUs is shown. Q, quarter.




decline in the baseline rate with 80% power at a 0.05 signifi-cance level, a sample size of 493,900 NICU central line–days would have been required—more than 16 times greater than the actual number of NICU central line–days observed during the 25-month control phase of this study (N = 29,601). Be-cause the study was so underpowered to detect differences in NICU CLABSI rates, no further analyses of this outcome were undertaken.

The main results of the analyses of hand hygiene compliance, adult ICU CLABSI, and VAP rates are summarized in Table 1 (above). Poisson regression analysis demonstrated a significant negative rate of change for VAP rates over time (slope = –0.049, p = 0.011) throughout the study period. The Poisson hierar-chical mixed model analysis, controlling for the correlation of outcome rates within hospitals, demonstrated a significant negative rate of change for adult ICU CLABSI rates over the study period (slope = –0.015, p = 0.020). In the separate anal-yses using hand hygiene compliance as the covariate, Poisson regression demonstrated a significant negative rate of change for VAP rates as hand hygiene compliance increased (slope = –2.25, p = 0.023). The hierarchical Poisson regression for adult ICU CLABSI rates demonstrated a similarly negative rate of change as hand hygiene compliance increased, but the p value fell just short of the 5% level of significance (slope = –1.19, p = 0.063).

DiscussionIn the quest to reduce HAIs, achieving and sustaining high lev-els of hand hygiene compliance have proved elusive goals for health care organizations, despite the high priority placed on them by organizations such as the World Health Organization and the CDC.27,28 MHHS used the TST29,30 to improve hand

hygiene across its 12 hospitals. We evaluated the extent of im-provement in hand hygiene compliance and the temporal asso-ciation between that improvement and the rates of adult ICU and NICU CLABSI and VAP. Because one of MHHS’s hospi-tals had participated in the Joint Commission Center for Trans-forming Healthcare’s original hand hygiene improvement proj-ect,13 had implemented hand hygiene improvement interven-tions throughout the hospital, and had increased compliance before the systemwide implementation of the TST, we limited this evaluation to MHHS’s other 11 hospitals.

Using the TST, MHHS hospitals increased hand hygiene compliance from a baseline of 58.1% to 95.6% by the last 12 months of the control phase, a relative improvement of 65% (p < 0.0001).

These results compare favorably to the Center’s original proj-ect, on which the development of the TST was based.13 In that project, eight hospitals improved hand hygiene compliance from 47.5% to 81.0% (a relative increase of 70.5%) and sustained those improvements over the 11-month control phase of that project. One previous evaluation of the TST in 174 health care organizations demonstrated an increase in hand hygiene com-pliance from a baseline of 57.9% to 83.5% during the improve phase.14 That study did not collect control phase data to assess for how long improvements were sustained. In the present study, a similar level of compliance was achieved during its improve phase (84.4%), but a higher level was achieved in the control phase and sustained for 25 months. The present study also demonstrated that MHHS achieved another important goal of the TST’s con-trol phase: reducing variability among its hospitals around a high level of average performance in hand hygiene compliance. Figure 3 shows that reduced variability visually, and the data anal ysis

Table 1. Main Results by Study Phase

Study phase Dates

No. of months

Hand hygiene

compliance (%)

No. of observations

(000s)

P value vs.

baseline

Adult ICU CLABSI rate (per

1,000 line- days)

Line-days

(000s)

P value vs.

baseline

VAP rate* (per 1,000

vent days)

Vent-days

(000s)

P value vs.

baselineBaseline Oct 2010–

May 2011 8 58.1 31.6 — 0.83 29.1 — 1.04 22.1 —

Improve June 2011–Nov 2012

18 84.4 145.1 < 0.001 0.63 70.5 0.298 0.57 45.8 0.045

Control 1 Dec 2012–Dec 2013

13 94.7 121.3 < 0.001 0.58 56.1 0.196 NA† NA NA

Control 2 Jan 2014–Dec 2014

12 95.6 110.1 < 0.001 0.42 49.7 0.024 NA NA NA

CLABSI, central line–associated bloodstream infection; VAP, ventilator-associated pneumonia; vent, ventilator; NA, not available. * Improve period for VAP is June 2011 through December 2012.† Data for VAP not available after December 2012 because of change in definition. (See text).




demonstrated that the reduction was statistically significant.This study also evaluated the association between rising

rates of hand hygiene compliance and rates of two important HAIs: adult ICU CLABSI and VAP. We used Poisson regres-sion analysis to assess the association of each of these outcomes in three different ways, using three different independent vari-ables: study phase (baseline versus improve [VAP] or control [CLABSI]), the 51 months of the study period, and monthly hand hygiene compliance. These analyses produced six tests of association. Five of the six were statistically significant (study phase for both CLABSI and VAP, time in months during the study period for CLABSI and VAP, and monthly hand hygiene compliance for VAP). The sixth test (monthly hand hygiene compliance for CLABSI) was in the hypothesized direction of lower CLABSI rates with higher rates of compliance, but the p value of 0.063 did not meet the conventional 5% threshold of significance. We believe that these data provide strong evidence that MHHS’s success in achieving and sustaining high levels of hand hygiene compliance likely played a significant role in reducing the rates of adult ICU CLABSI and VAP across the hospitals in its system.

Of further note, the baseline rate of adult ICU CLABSI was 0.83 per 1,000 line-days, which is considerably lower, for exam-ple, than the postintervention mean of 1.4 achieved in Michigan by 70 adult ICUs at 15 to 18 months of follow-up in the Key-stone project.31 It is also substantially lower than the 2009 United States national average of 1.65, as estimated by the CDC.32 Sim-ilarly, the baseline VAP rate of 1.04 is lower than the CDC–esti-mated 2009 national rates for medical ICUs in major teaching (1.9) or in other hospitals (1.4), in surgical ICUs (3.8), or in combined medical/surgical ICUs (2.0 in major teaching hospi-tals, 1.4 in small [≤ 15 beds] units or 1.2 in larger [> 15 beds] units in other hospitals).33 We believe that these low baseline rates were likely due to MHHS’s previous implementation of the CLABSI and VAP bundles.25,26 The data in Figures 4, 5, and 6 are consistent with but do not by themselves provide a ro-bust test of this hypothesis. However, we believe that any effect of the previous implementation of the bundles on ICU CLAB-SI or VAP rates took place well before the hand hygiene im-provement initiative. The aggregate data on rates of adult ICU CLABSI and VAP showed no time trend in the 12 months be-fore the hand hygiene TST deployment began.

The data from this study suggest that procedure-specific processes to reduce infections related to devices such as central lines or ventilators or others, such as urinary catheters, may not achieve extremely low rates of infection unless they are accom-panied by high rates of hand hygiene compliance.

criTicaL SucceSS facTorS

Several critical success factors likely contributed to the sig-nificant increase in hand hygiene compliance during MHHS’s TST RPI project. Leadership support and oversight were key to success because leaders allocated the necessary resources to complete the project, which was part of a larger initiative to be-coming an HRO. Also key, the RPI project was led by an MBB seasoned in RPI methodology.

Leadership support enabled MHHS to spend critical time in each step of the TST, beginning with the first four steps of the TST process, “Getting Started,” “Training Secret Observers and Just-in-Time (JIT) Coaches,” “Measuring Baseline Com-pliance,” and “Determining Factors,” which mirror the Define, Measure, and Analyze phases of the Six Sigma DMAIC process. By spending adequate time in these initial phases before under-taking improvement interventions, staff were shown the value of accurate and robust data collection so that solutions would be targeted to address root causes and optimize the probability of hardwiring the improvement. Because of the perceived ur-gency of identifying and addressing problems, health care or-ganizations conducting process improvement projects may rush through or entirely skip the first three phases of DMAIC and begin implementing improvements prematurely. By doing so, these organizations may not be able to completely define their problems nor collect robust data to accurately measure compli-ance or identify contributing factors. In these instances, root causes may not effectively be addressed, and the same problems will likely continue to appear.34,35

Also central to the project’s success was the TST method-ology and Web-based application, which guided MHHS on how to collect, input, and analyze reliable data from a sam-pling of observations provided by secret observers. When mea-suring hand hygiene compliance in the past, MHHS estimated its baseline compliance between 80% and 95%. Through the TST, MHHS accurately measured a compliance rate of 58.1% (95% CI, 42.7% to 72.1%). When estimated using the TST methodology, compliance was statistically significantly low-er than previously estimated using known observers. In addi-tion, implementing proven solutions targeted at the respective unit-level factors, as opposed to systemwide generic and stan-dardized solutions,13 contributed to the interventions’ effec-tiveness and sustainability. Compliance rates continue to be monitored monthly using the TST and are reported throughout the organization, both contributing to the sustainability of the improvements. The hand hygiene measurement and improve-ment process is now considered “standard work”36,37—that is, work that is standard across the system—and will be continued




indefinitely. In addition, significant gains in hand hygiene com-pliance early in the project became motivating factors in their own right. As implementation matured, the sustained increase in hand hygiene compliance was associated temporally with a decrease in HAI rates throughout the system.

LiMiTaTionS

This study has several limitations. The TST was not imple-mented using a randomized controlled design. Factors other than the interventions derived from the TST may have con-tributed to the observed increases in hand hygiene compliance. Although the training of the secret observers who collected data on hand hygiene compliance was standardized and maintained throughout the 51 months of the project, we did not formal-ly assess interrater reliability among the observers. This study demonstrated a prolonged (25-month) period of sustained high levels of compliance following the deployment of TST–mediated interventions. The TST is designed to be used by organizations with no expertise in RPI.14 However, MHHS has a strong RPI program, and its TST implementation was led by an MBB. Although 174 initial TST users demonstrated a level of improve-ment from baseline to the improve phase comparable to that reported here,14 we do not know whether organizations with-out RPI expertise would experience the same level of sustained improvement (in the control phase) as demonstrated by MHHS. Finally, although we observed a substantial decrease in adult ICU CLABSI and VAP rates in temporal association with the increases in hand hygiene compliance, factors other than im-proved hand hygiene (for example, the Certified Zero Award recognition program) may have contributed to the reduction in these HAIs.

ConclusionUsing the steps and tools embedded in the TST for hand hy-giene, from a baseline level of 58.1%, MHHS achieved a 65% relative improvement in hand hygiene compliance across 11 hospitals. A high level of compliance (94.7%) was achieved in the first 13 months of the control phase and sustained at 95.6% for the last 12 months of the study period. Baseline rates of adult ICU CLABSI and VAP were relatively low (0.83 per 1,000 line-days and 1.04 per 1,000 ventilator-days, respective-ly). In temporal association with the increases in hand hygiene compliance, these HAI rates decreased further, by relative rates of 49% and 45%, respectively. J

References1. Shabot MM, et al. Memorial Hermann: High reliability from board to bed-side. Jt Comm J Qual Patient Saf. 2013;39:253–257.2. Chassin MR, Loeb JM. The ongoing quality improvement journey: Next stop, high reliability. Health Aff (Millwood). 2011;30:559–568.3. Wolterman D, Shabot MM: A new standard: Aim for safety of planes, nuclear plants. Mod Healthc. 2011 Aug 1;41(31):27.4. Chassin MR, Loeb JM. High-reliability healthcare: Getting there from here. Milbank Q. 2013;91:459–490.5. Joint Commission Center for Transforming Healthcare. Robust Process Improvement®. Accessed Nov 29, 2015. http://www.centerfortransforming healthcare.org/about/rpi.aspx.6. Centers for Disease Control and Prevention. Hand Hygiene in Health-care Settings: Guidelines. (Updated: May 19, 2011.) Accessed Nov 29, 2015. http://www.cdc.gov/handhygiene/Guidelines.html.7. Joint Commission Center for Transforming Healthcare. Facts About the Hand Hygiene Project. Dec 2014. Accessed Nov 29, 2015. htttp://www .centerfortransforminghealthcare.org/projects/detail.aspx?Project=3.8. World Health Organization. Clean Care Is Safer Care. Accessed Nov 29, 2015. http://www.who.int/gpsc/5may/en/.9. Pittet D, et al. Effectiveness of a hospital-wide programme to improve com-pliance with hand hygiene. Lancet. 2000 Oct 14;356:1307–1312.10. Rosenthal VD, Guzman S, Safdar N. Reduction in nosocomial infection with improved hand hygiene in intensive care units of a tertiary care hospital in Argentina. Am J Infect Control. 2005;33:392–397.11. Larson E, Kretzer EK. Complaince with handwashing and barrier precau-tions. J Hosp Infect. 1995;30 Suppl:88–106.12. Huis A, et al. A systematic review of hand hygiene improvement strategies: A behavioural approach. Implement Sci. 2012 Sep 14;7:92.

M. Michael Shabot, MD, FACS, FCCM, FACMI, is System Exec-utive Vice President and Chief Clinical Officer, Memorial Hermann Health System, Houston. Mark R. Chassin, MD, FACP, MPP, MPH, is President and CEO, The Joint Commission, Oakbrook Terrace, Illinois. Anne-Claire France, PhD, CPHQ, MBB, FACHE, is Pro-cess Improvement Specialist, Enterprise Quality, Patient Safety, and Infection Control, Memorial Hermann Health System; President and Owner, Houston Health Innovations, LLC (HHI); and a member of The Joint Commission Journal on Quality and Patient Safety’s Edito-rial Advisory Board. Juan Inurria, FACHE, FABC, CPHQ, is System Executive, Memorial Hermann Health System. Jan Kendrick, MA, is Director of Business Operations, and Stephen P. Schmaltz, PhD, is Associate Director, Center for Data Management and Analysis, The Joint Commission. Please address correspondence to Michael Shabot, [email protected].

Online Only Contenthttp://www.ingentaconnect.com/content/jcaho/jcjqs

See the online version of this article for Appendix 1. Time Line for Implementation of the Targeted

Solutions Tool® (TST®) at Memorial Hermann Health System (MHHS)

Appendix 2. Hand Hygiene Data Collection Form




13. Chassin MR, Mayer C, Nether K: Improving hand hygiene at eight hos-pitals in the United States by targeting specific causes of noncompliance. Jt Comm J Qual Patient Saf. 2015;41:4–12.14. Chassin MR, et al. Beyond the collaborative: Spreading effective improve-ment in hand hygiene compliance. Jt Comm J Qual Patient Saf. 2015;41: 13–25.15. Joint Commission Center for Transforming Healthcare. Project Detail: Hand Hygiene Project. Accessed Nov 29, 2015. http://www.centerfortrans-forminghealthcare.org/projects/detail.aspx?Project=3.16. Joint Commission Center for Transforming Healthcare. FAQs: Hand Hygiene TST®. Accessed Nov 29, 2015. http://www.centerfortransforming healthcare.org/faq.aspx?CategoryId=49#1111.17. The Joint Commission. Measuring Hand Hygiene Adherence: Overcoming the Challenges. Oak Brook, IL: Joint Commission Resources, 2009. Accessed Nov 29, 2015. http://www.jointcommission.org/assets/1/18/hh_monograph.pdf.18. Kohli E, et al. The effect of recognized observers on measurement of hand hygiene compliance in high and low performing inpatient units (abstract). Paper presented at the Annual Meeting of the Society for Healthcare Epidemi-ology of America, Orlando, FL, Apr 7, 2008.19. Gould DJ, et al. Measuring hand washing performance in health service audits and research studies. J Hosp Infect. 2007;66:109–115.20. Centers for Disease Control and Prevention. Outline for Healthcare-Asso-ciated Infections Surveillance. Apr 2006. Accessed Nov 29, 2015. http://www .cdc.gov/nhsn/PDFS/OutlineForHAISurveillance.pdf.21. Gebski V, et al. Modelling interrupted time series to evaluate prevention and control of infection in healthcare. Epidemiol Infect. 2012;140:2131–2141.22. Cameron AC, Trivedi PK. Regression Analysis of Count Data. Cambridge, UK: Cambridge University Press, 1998.23. Molenberghs G, Verbeke G. Models for Discrete Longitudinal Data. New York City: Springer Science+Business Media, 2005.24. Tukey JW. Exploratory Data Analysis. Reading, MA: Addison-Wesley, 1977. 25. The Joint Commission. Preventing Central Line–Associated Infections: Use-ful Tools, An International Perspective. CVC Insertion Bundles. Nov 20, 2013. Accessed Nov 29, 2015. http://www.jointcommission.org/assets/1/6/CLABSI _Toolkit_Tool_3-18_CVC_Insertion_Bundles.pdf.26. Munro N, Ruggiero M. Ventilator-associated pneumonia bundle: Recon-

struction for best care. AACN Adv Crit Care. 2014;25:163–175.27. World Health Organization (WHO). WHO Guidelines on Hand Hygiene in Health Care. Geneva: WHO, 2009. Accessed Nov 29, 2015. http://apps.who .int/iris/bitstream/10665/44102/1/9789241597906_eng.pdf.28. Boyce JM, Pittet D; Healthcare Infection Control Practices Advisory Com-mittee; HICPAC/SHEA/APIC/IDSA Hand Hygiene Task Force. Guideline for Hand Hygiene in Health-Care Settings. Recommendations of the Healthcare Infection Control Practices Advisory Committee and the HICPAC/SHEA/APIC/IDSA Hand Hygiene Task Force. Society for Healthcare Epidemiology of America/Association for Professionals in Infection Control/Infectious Dis-eases Society of America. MMWR Recomm Rep. 2002 Oct 25;5(RR-16):1–45.29. Joint Commission Center for Transforming Healthcare. Targeted Solu-tions Tool for Hand Hygiene. Accessed Nov 29, 2015. http://www.centerfor transforminghealthcare.org/tst_hh.aspx.30. Joint Commission Center for Transforming Healthcare. Facts About the Targeted Solutions Tool® (TST®). Aug 2015. Accessed Nov 29, 2015. http://www.centerfortransforminghealthcare.org/assets/4/6/TST_Fact_Sheet.pdf.31. Pronovost P, et al. An intervention to decrease catheter-related bloodstream infections in the ICU. N Engl J Med. 2006 Dec 28;355:2725–2732. Erratum in: N Engl J Med. 2007 Jun 21;356:2660.32. Srinivasan A, et al.; Centers for Disease Control and Infection. Vital signs: Central line-associated blood stream infections—United States, 2001, 2008 and 2009. MMWR Morb Moral Wkly Rep. 2011 Mar 4;60(8):243–248.33. Centers for Disease Control and Prevention. National Healthcare Safety Network (NHSN) Report, Data Summary for 2009, Device-Associated Mod-ule. Dudeck MA, et al. 2010. Accessed Nov 29, 2015 http://www.cdc.gov /nhsn/PDFs/dataStat/2010NHSNReport.pdf.34. Arthur J. Lean Six Sigma for Hospitals: Simple Steps to Fast, Affordable, and Flawless Healthcare. New York City: McGraw-Hill, 2011.35. Futerer SL. Lean Six Sigma Case Studies in the Healthcare Enterprise. New York City: Springer, 2014.36. Barnas K. ThedaCare’s business performance system: Sustaining continu-ous daily improvement through hospital management in a lean environment. Jt Comm J Qual Patient Saf. 2011;37:387–399.37. Womack JP, Jones DT. Lean Thinking: Banish Waste and Create Wealth in Your Corporation, 2nd ed. New York City: Free Press, 2003.



Volume 42 Number 1January 2016 AP1

Online Only ContentAppendix 1. Time Line for Implementation of the Targeted Solutions Tool® (TST®) at

Memorial Hermann Health System (MHHS)

(continued on page AP2)

Lead

Re

spon

sibilit

yJu

ly20

10Au

g20

10Se

pt20

10Oc

t20

10No

v 20

10De

c20

10Ja

n 20

11

Feb

2011

Marc

h20

11Ap

ril

2011

May

2011

June

20

11

Seni

or L

eade

rshi

p Al

ignm

ent &

Res

ourc

ing

1.D

evel

op s

yste

m h

and

hygi

ene

polic

yG

inny

K

enne

dy/Q

PS

2.C

omm

unic

ate

hand

hyg

iene

goa

l for

FY

11 to

site

sw

hich

is:

Est

ablis

h re

liabl

e m

easu

rem

ent

syst

eman

d id

entif

y co

ntrib

utin

g fa

ctor

s by

uni

t. F

Y12

will

focu

s on

targ

eted

sol

utio

ns b

ased

on

FY11

dat

a.

Dr.

Sha

bot/J

uan

Inur

ria/Q

PS

3.

Prov

ide

site

ove

rvie

w o

f han

d hy

gien

e pl

an to

C

NOs,

Qua

lity

Dirs

, Qua

lity

Pt S

afet

y C

ounc

ilA

nne-

Cla

ire

Fran

ce/Q

PS

Si

te-S

peci

fic W

ork

Befo

re C

TH

Web

Por

tal G

o-Li

ve

4.Id

entif

y fa

cilit

y-ba

sed

site

han

d hy

gien

eim

plem

enta

tion

team

reso

urce

s:•

Exe

cutiv

e S

pons

or (C

NO

)•

Faci

lity

Pro

cess

Ow

ner

Site

Le

ader

ship

5.If

ther

e is

a n

eed

to s

tagg

er g

o-liv

es, d

evis

e ho

wto

div

ide

all p

t car

e un

its in

to th

ree

roun

ds.

6.Id

entif

y R

ound

ON

E P

artic

ipat

ing

Uni

ts a

ndR

ound

ON

E P

roce

ss O

wne

r

Site

Le

ader

ship

7.As

requ

este

d, s

ite-s

peci

fic o

rient

atio

n to

the

CTH

met

hodo

logy

is p

rovi

ded

to e

ach

faci

lity

base

dim

plem

enta

tion

team

incl

udin

g E

xecu

tive

Spo

nsor

,P

roce

ss O

wne

rs, Q

ualit

y D

irect

ors,

and

ICP

s.

Ann

e-C

laire

Fr

ance

/QP

S

8.B

uild

a te

am o

f 3-7

in e

ach

roun

d O

NE

parti

cipa

ting

unit

Rou

nd O

ne

Pro

cess

O

wne

rs

9.Id

entif

y 2-

3 se

cret

obs

erve

rs/d

ata

colle

ctor

s pe

rro

und

one

parti

cipa

ting

unit

(EV

S, D

ieta

ry, P

lant

Ops

, Cha

plai

ns, m

ake

grea

t sec

ret o

bser

vers

)

Pro

cess

O

wne

r & U

nit

Team

s10

.Ide

ntify

1-2

JIT

coa

ches

per

roun

d O

NE

parti

cipa

ting

unit

Pro

cess

O

wne

r & U

nit

Team

s



Volume 42 Number 1January 2016AP2


Memorial Hermann Health System (MHHS) (continued)

(continued on page AP3)

Lead

Re

spon

sibilit

yJu

ly20

10Au

g20

10Se

pt20

10Oc

t20

10No

v 20

10De

c20

10Ja

n 20

11

Feb

2011

Marc

h20

11Ap

ril

2011

May

2011

June

20

11

CTH

Por

tal G

o-Li

ve S

ept 1

2 11

.Ide

ntify

who

will

hav

e ac

cess

to W

eb-b

ased

CTH

Por

tal

Pro

cess

O

wne

r & U

nit

Team

s

12.A

cces

s an

d Pl

ay w

ith C

TH P

orta

lA

ll id

entif

ied

to A

cces

s 13

.Rou

nd O

NE

secr

et o

bser

vers

trai

ned

and

certi

fied

usin

g po

rtal r

esou

rces

Site

Tea

m

14.I

dent

ify R

ound

TW

O p

artic

ipat

ing

units

and

Pro

cess

O

wne

rsS

ite

Lead

ersh

ip

15.R

ound

ON

E se

cret

obs

erve

rs in

itiat

e ba

selin

e da

taco

llect

ion

on p

artic

ipat

ing

units

– u

se p

orta

lre

sour

ces

(col

lect

2 w

eeks

of d

ata

and

lock

inpr

oces

s st

abilit

y)

Site

Tea

m

16.R

ound

ON

E S

ecre

t Obs

erve

rs c

ontin

ue to

col

lect

com

plia

nce

data

Site

Tea

m

17.I

dent

ify R

ound

TW

O S

ecre

t Obs

erve

rsS

ite

Lead

ersh

ip

18.L

ock

in b

asel

ine

data

refle

ctin

g st

able

pro

cess

for

Rou

nd O

NE

– us

e po

rtal r

esou

rces

. If

data

not

stab

le th

en a

ddre

ss s

peci

al c

ause

s an

d co

llect

add

lda

ta.

Site

Te

am/Q

PS

19.T

rain

Rou

nd T

WO

Sec

ret O

bser

vers

usi

ng W

ebpo

rtal

Site

Tea

m

20.I

dent

ify R

ound

TH

RE

E pa

rtici

patin

g un

its a

ndpr

oces

s ow

ners

Site

Tea

m

Iden

tify

Con

tribu

ting

Fact

ors

21.T

rain

roun

d O

NE

just

-in-ti

me

(JIT

) coa

ches

toco

llect

con

tribu

ting

fact

ors

– us

e W

eb p

orta

lS

ite T

eam

22.

Iden

tify

Rou

nd T

HR

EE

Sec

ret O

bser

vers

Site

Tea

m23

.R

ound

TW

O s

ecre

t obs

erve

rs in

itiat

e ba

selin

e da

taco

llect

ion

on p

artic

ipat

ing

units

– u

se p

orta

lre

sour

ces

(col

lect

2 w

eeks

of d

ata

and

lock

inpr

oces

s st

abilit

y)

Site

Tea

m



Volume 42 Number 1January 2016 AP3


Memorial Hermann Health System (MHHS) (continued)

QPS, Quality and Patient Safety; FY, fiscal year; CNO, chief nursing officer; Dirs, directors; Pt, patient; CTH, Joint Commission Center for Transforming Healthcare; ICP, infection control preventionist; EVS, environmental services; Ops, operations; JIT, just-in-time; addl, additional.

Lead

Re

spon

sibilit

yJu

ly20

10Au

g20

10Se

pt20

10Oc

t20

10No

v 20

10De

c20

10Ja

n 20

11

Feb

2011

Marc

h20

11Ap

ril

2011

May

2011

June

20

11

24.R

ound

TW

O s

ecre

t obs

erve

rs c

ontin

ue to

col

lect

com

plia

nce

data

Site

Tea

m

25.R

ound

ON

E JI

T co

ache

s co

llect

con

tribu

ting

fact

ors

fo

r non

-com

plia

nce.

Use

por

tal r

esou

rces

Site

Tea

m

26.T

rain

Rou

nd T

HR

EE

Sec

ret O

bser

vers

usi

ng W

ebpo

rtal

Site

Tea

m

27.I

dent

ify c

ontri

butin

g fa

ctor

s fo

r Rou

nd O

NE

Uni

ts –

use

porta

l res

ourc

esS

ite T

eam

28. T

rain

Rou

nd T

WO

JIT

coa

ches

to c

olle

ctco

ntrib

utin

g fa

ctor

sS

ite T

eam

29.R

ound

TW

O J

IT c

oach

es c

olle

ct c

ontri

butin

gfa

ctor

s fo

r non

-com

plia

nce.

Use

por

tal r

esou

rces

Site

Tea

m

30.I

dent

ify c

ontri

butin

g fa

ctor

s fo

r Rou

nd T

WO

Uni

ts –

use

porta

l res

ourc

esS

ite T

eam

31.

Rou

nd T

HR

EE

secr

et o

bser

vers

initi

ate

base

line

data

col

lect

ion

on p

artic

ipat

ing

units

– u

se p

orta

lre

sour

ces

(col

lect

2 w

eeks

of d

ata

and

lock

inpr

oces

s st

abilit

y)

Site

Tea

m

32.R

ound

TH

RE

E se

cret

obs

erve

rs c

ontin

ue to

col

lect

com

plia

nce

data

Site

Tea

m

33.T

rain

roun

d TH

RE

E J

IT c

oach

es to

col

lect

cont

ribut

ing

data

34.R

ound

TH

RE

E JI

T co

ache

s co

llect

con

tribu

ting

fact

ors

for n

on-c

ompl

ianc

e. U

se p

orta

l res

ourc

esS

ite T

eam

35.I

dent

ify c

ontri

butin

g fa

ctor

s fo

r Rou

nd T

HR

EE

Uni

ts–

use

porta

l res

ourc

esS

ite T

eam



Volume 42 Number 1January 2016AP4

Online Only ContentAppendix 2. Hand Hygiene Data Collection Form

1. Us

e a se

para

te ro

w fo

r eac

h en

try o

r exit

.2.

Whe

n th

ere i

s a d

efec

t (wa

sh in

/out

=no)

, che

ck an

y app

licab

le ob

serv

ed co

ntrib

utin

g fa

ctor

.3.

The “

obse

rved

by a

skin

g” se

ctio

n is

for J

IT co

ache

s onl

y.4.

Emer

genc

y situ

atio

ns ar

e EXC

LUDE

D fro

m th

e dat

a col

lectio

n pr

oces

s.

Colle

cted b

y:Ro

le (ci

rcle o

ne):

Obse

rver

Coac

hUn

it:

Non

Obse

rvab

le

Date

of ob

serva

tions

:Ha

nd H

ygien

e Obs

erva

tion

and

Cont

ribut

ing

Fact

or F

orm

Poss

ible

Cont

ribut

ing

Fact

ors t

o W

ashi

ng

Obse

rvab

le

1 2 3 4 5 6 7 8 9 10Inst

ruct

ions

:

12

34

56

78

910

1112

1314

1516

1718

1920

2122

Dispen

ser lo

catio

n

Did pe

rson w

ash?

Entry o

r exit

?

Observ

ation

Num

ber

Check

box i

f obs

erved

durin

g rou

nds

RN = Nurs

eNA = N

ursing

Assista

nt

MD = Doct

orRT = R

epirato

ry Ther

apy

PT = Phys

ical Ther

apy

Diet = D

ietary T

echnic

ian

Lab = L

ab Tec

hnicia

n

HSK = Hous

ekeepi

ng

all oth

ers ple

ase ide

ntify in

comment

s

Enter h

our o

f obs

ervati

on

in

24-ho

ur (m

ilitary

) time

Dispen

ser e

mpty

Dispen

ser b

roken

Equipm

ent s

hared

Hands

full s

uppli

es

Hands

full m

eds

Impro

per u

se of

glov

es

Follow

perso

n entr

y or e

xit

Freque

nt en

try or

exit

Admiss

ions o

r disc

harge

s proc

ess

Isolat

ion ar

ea (g

own +

glov

es)

Lack

of im

mediat

e fee

dbac

k

Distrac

ted or

forgo

t

Percep

tion H

H not re

quire

d

Percep

tion o

f skin

irrita

tion

Other c

ontrib

uting

facto

r

Com

men

ts

EN E

X

EN E

X

EN E

X

EN E

X

EN E

X

EN E

X

EN E

X

EN E

X

EN E

X

EN E

X

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

6.D

ispe

nser

loca

tion

is n

ot in

pat

h of

per

son

or is

obs

truc

ted

or h

idde

n7.

Dis

pens

er is

em

pty

8.D

ispe

nser

is b

roke

n9.

Equi

pmen

t sha

red

or d

ispo

sal a

rea

(use

of e

quip

men

t sha

red

betw

een

patie

nts

(i.e.

, vita

l sig

n m

achi

ne, p

orta

ble

x-ra

y, e

tc.)

10.

Han

ds fu

ll: s

uppl

ies

or e

quip

men

t (e.

g., f

ood

tray

s, la

b su

pplie

s)11

.H

ands

full

med

s12

.G

love

s (e

.g.,

impr

oper

use

of o

r not

was

hing

bef

ore

or a

fter p

uttin

g gl

oves

on

or o

ff)13

.Pe

rson

ent

erin

g or

exi

ting

follo

wed

som

eone

who

did

not

was

h14

.Fr

eque

nt e

ntry

and

exi

t of p

atie

nt a

rea

15.

Adm

issi

ons

or d

isch

arge

pro

cess

16.

Isol

atio

n ar

ea (g

own

+ gl

oves

whe

n re

quire

d)17

.La

ck o

f im

med

iate

feed

back

to p

erso

n fo

r han

d hy

gien

e co

mpl

ianc

e18

.D

istr

actio

ns/fo

rget

s/la

ck o

f kno

wle

dge/

chos

e no

t to

was

h19

.Pe

rcep

tion

that

if n

othi

ng is

touc

hed

in th

e pa

tient

car

e ar

ea h

and

hygi

ene

isno

t nec

essa

ry20

.Pe

rcep

tion

of s

kin

irrita

tion

or d

islik

e of

alc

ohol

-bas

ed h

and

rub

21.

Oth

er

Con

trib

utin

g Fa

ctor

s

Circ

le ro

le o

f hea

lth c

are

prof

essi

onal

obs

erve

d

RN N

A M

D R

T P

T D

iet L

ab H

SK

CM / S

W

Phar

m R

ad O

ther

CM / SW = C

ase Mgm

t / Soci

al Work

er

Pharm = P

harmaci

st

Rad = R

adiolo

gy Tec

h

RN N

A M

D R

T P

T D

iet L

ab H

SK

CM / S

W

Phar

m R

ad O

ther

RN N

A M

D R

T P

T D

iet L

ab H

SK

CM / S

W

Phar

m R

ad O

ther

RN N

A M

D R

T P

T D

iet L

ab H

SK

CM / S

W

Phar

m R

ad O

ther

RN N

A M

D R

T P

T D

iet L

ab H

SK

CM / S

W

Phar

m R

ad O

ther

RN N

A M

D R

T P

T D

iet L

ab H

SK

CM / S

W

Phar

m R

ad O

ther

RN N

A M

D R

T P

T D

iet L

ab H

SK

CM / S

W

Phar

m R

ad O

ther

RN N

A M

D R

T P

T D

iet L

ab H

SK

CM / S

W

Phar

m R

ad O

ther

RN N

A M

D R

T P

T D

iet L

ab H

SK

CM / S

W

Phar

m R

ad O

ther

RN N

A M

D R

T P

T D

iet L

ab H

SK

CM / S

W

Phar

m R

ad O

ther


Documents

Improvement from January 2016 Front Office to Front Line …€¦ · spreading and sustaining improved compliance have proved dif-ficult.11,12 As part of its efforts to reduce HAIs,